L

LLMs

💬 Epic prompts to turbo-charge your LLM chatbots.

🤖 AI chat & search summaries in Google Search, powered by the latest LLMs

A unified Python interface to select and use multiple Large Language Model (LLM) providers through a common API.

🛒 AI chat & product/category summaries in Amazon shopping, powered by the latest LLMs

Kevlar Benchmark: OWASP Top 10 for Agentic Apps (AI-Agents) 2026 a Red Team Benchmark.

Model Verification Layer is a modular system for evaluating, comparing, and validating the behavior of large language models using structured benchmarks, logic consistency checks, cross-model consensus analysis, and policy-aware constraints. It provides a transparent framework for understanding how different AI models perform under identical conditions, enabling more reliable model selection, safer deployment, and user-adaptive decision making. https://roxanneardary.com/model-verification-layer/

POS CLI is a terminal-based Point of Sale application, designed as a tool for AI Agents (LLMs). Lightweight, fast, single binary, no external dependencies.

KiM Explorer is a two-stage RAG application for transport policy research publications from the KiM Netherlands Institute for Transport Policy Analysis. Users perform semantic search to identify relevant documents, manually select publications, then interact with an LLM using full document context rather than chunks. Built with Python/NiceGUI/OpenAI API, featuring citation generation, conversation history, filtering, and web/CLI interfaces. https://explorer.kim.rijkscloud.nl/

A diagnostic framework for measuring LLM vulnerability to Affective Contextual Erosion (ACE) and related liminal attack vectors. Delirium is not an exploitation tool. It is a standardized benchmark designed to detect the precise moment when a language model's attention weights shift from serving a system prompt to serving an emergent interpersonal pattern — before harm occurs.

Securekit is a protocol-agnostic security kernel that enforces zero-trust, sandboxed execution for AI tool use. It sits between any LLM or agent system and its tools, validating, isolating, and auditing every action to prevent unsafe execution across MCP, OpenAI tools, and custom AI protocols. https://roxanneardary.com/securekit/

SynapCache is a distributed, zero-memory-loss caching system for all LLM outputs, designed to be LLM-agnostic, high-performance, and scalable. It combines TurboQuant-style compression, semantic search, edge caching, and predictive intelligence to store and retrieve outputs efficiently, ensuring instant recall and minimal resource usage. With developer-friendly SDKs, plugin support, and multi-cloud deployment, SynapCache provides a universal neural memory layer for modern AI workflows. https://roxanneardary.com/synapcache/

AI assistant for the GNOME Evolution mail client — summarize incoming messages and draft replies with your choice of LLM.

Paper: "A Comprehensive Evaluation of Pre-Trained Language Models for Irony Detection in Tweets"

LLM Workflow Router is a stateless middleware engine designed to enforce explicit execution topology in AI systems that rely on large language models. It evaluates interaction metadata against strictly declared workflow rules and returns a terminal state.

GeoListing is an AGPL 3.0+ licensed modular AI platform for real estate marketing, semantic search optimization, compliance automation, and AI-powered property discovery. Built for brokerages, agents, developers, and proptech platforms, GeoListing combines listing generation, SEO and LLM optimization, legal intelligence, market analytics, and advanced publishing workflows into open semantic infrastructure for modern real estate ecosystems. https://roxanneardary.com/geolisting/

Red Team AI Benchmark: Evaluating LLMs for authorized offensive-security tasks. Red Team AI Benchmark is a CLI model-evaluation benchmark. It measures how LLMs understand and respond to red-team questions and security scenarios; it is not a tool for carrying out those activities. Version 2 uses a rubric-based dataset instead of judging answers only against one golden response.

This project is a Retrieval-Augmented Generation (RAG) application built using LangChain. It leverages advanced language models and vector databases to answer questions about epidemiological modeling, software development, and maintaining the EPP model for HIV modeling.

OptiRank is an open-source SEO and LLM optimization tool that analyzes existing website content and provides actionable recommendations to improve search rankings, semantic clarity, and AI discoverability. Built with a modular Python backend, React dashboard, and NLP/LLM-powered analysis layer, it helps developers and site owners optimize pages for both traditional search engines and modern AI-driven search systems.

A web AI interface using a selected number of models and trained to teach the German language.

AI Image Renamer is a Python CLI tool that leverages generative artificial intelligence to rename your images based on what the image shows.